Project Summary A vaccine that elicits broadly neutralizing antibodies (bnAb) could protect against HIV/AIDS. All bnAbs must act on the membrane embedded form of the HIV-1 envelope glycoprotein (m-Env), however typical vaccines use soluble, truncated forms of Env as an imperfect surrogate that is simpler to formulate. We have been developing a m-Env vaccine strategy that presents purified m-Env in multivalent form on liposomes, which we term, m-Env liposomes (MELs). MELs have been enabled in part by development of stable, well- ordered m-Env and producer cells that generate them in high yield. Preliminary Studies show that heterologous MEL immunization elicited sporadic tier 2 cross neutralizing antibody responses. Immunogenicity of the transmembrane Env subunit, gp41, is underexplored in the membrane context. With MELs, immunogenicity of m-Env gp41 can now be studied with molecular precision, and without the neoepitopes associated with soluble Env. In SA1, we will immunize rabbits using MELs to determine how neutralizing antibody responses are affected by gp41 immunofocusing strategies, including the creation of gp41 `glycan holes', boosting with multivalent epitope-specific immunogens based on membrane-proximal external region (MPER) and fusion peptide, and by boosting sequentially with heterologous Envs. Later, human Ig locus knockin (KI) mice will be immunized with m-Envs, including those that bind tightly to an inferred germline precursor (iGL) of a novel bnAb that we show has capacity to directly bind to the MPER and neutralize HIV-1. In SA2, we will isolate MPER/gp41 bnAbs from Chinese donors samples using B cell sorting and next generation sequencing bioinformatics techniques. We will also interrogate the gp41- specific B cell response in MEL immunized hu Ig KI mice to determine whether MPER bnAb-like iGLs expanded. In SA3, we will screen Envs from donors in part on the basis of affinity for MPER bnAb and cognate precursors to understand the mechanism of binding, and to generate novel immunogens. Overall, we aim to understand how to elicit bnAbs to gp41 that recognize a well guarded, but nevertheless vulnerable surface at the base of the HIV-1 Env spike.